connect
Connect to an nREPL server. Example: (connect {:host "localhost" :port 1234})
LABSCO SUMMARY
It talks to the process you already have running, so application state survives the conversation.
This is a live connection, not a sandbox: eval_form runs inside the process holding your application's state, which is what separates it from asking a model to reason about the source. get_ns_vars is the read side and it is unusually complete — doc string, file, line and the value a var holds right now, which is the question normally answered by adding a println and restarting. The surface is three tools, so the boundaries are clear: nothing here edits files or manages a project.
What it is
A three-tool bridge to an already-running nREPL server: connect, evaluate, inspect a namespace.
What you get
- connect, taking host and port, to attach to an nREPL that is already up
- eval_form, evaluating Clojure in a named namespace or the current one — its examples cover defining and calling a function, reloading code, and syncing dependencies
- get_ns_vars, returning a namespace's public vars as a map where each entry carries metadata (doc string, line number, file path) and the var's current value
Requirements
A running nREPL server, and its host and port.
Setup effort
One command — npx nrepl-mcp-server
USE CASES
01
Trying a fix against the system that is already running
The session attaches to the REPL you have open, and a form is evaluated in the namespace it belongs to, so a change is tested against live state instead of a restart.
02
Finding out what a namespace exposes and what it currently holds
Public vars come back with their docstrings, line numbers and file paths alongside the value each one holds at that moment, which is state the source file cannot tell you.
03
Reloading after an edit without dropping the session
Reload and dependency-sync calls are evaluated in the REPL like any other form, so changed code and newly added dependencies land in the process that is already running.
3 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
eval_form
Evaluate Clojure code in a specific namespace or the current one. Examples:
- Define and call a function: {"code": "(defn greet [name] (str \"Hello, \" name \"!\"))(greet \"World\"))"}
- Reload code: {"code": "(clj-reload.core/reload)"}
- Evaluate in a specific namespace: {"code": "(clojure.repl.deps/sync-deps)", "ns": "user"}
get_ns_vars
Get all public vars (functions, values) in a namespace with their metadata and current values. Example:
- List main namespace vars: (get_ns_vars {:ns "main"})
Returns a map where keys are var names and values contain:
- :meta - Metadata including :doc string, :line number, :file path
- :value - Current value of the var
BEFORE YOU INSTALL
What to know first
Tools you get
3 tools when we connected.
Requires
npx on PATH
CONTEXT COST
Light Standing cost
Rated by the size of the 3 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Add the block to your client, or run the command directly.
Server setup
Configuration file
JSON
{
"mcpServers": {
"nrepl-mcp-server": {
"command": "npx",
"args": [
"nrepl-mcp-server"
]
}
}
}npx
TERMINAL
npx nrepl-mcp-server
